So I’m struggling with Screen to World again. Well actually my son is and since I solved it before in my game I thought I could help him. But man it’s not as easy as I hoped. I know I used orthographic camera before but we really want to stick with the updated defold built in camera functions.
And it works when you run it; however, if you resize the window this is where the issues start. The larger the window change the more offset the OS mouse is from the in-game cursor.
So my question is, how do we properly calculate and translate the OS cursor position to the in-game cursor position after resizing the window? I did try sending a message from the default render script and have that working but it still wasn’t working properly in my screen to world function.
This is pretty basic stuff, I was hoping defold had a sample out there that would make this so much easier. If one exists, please let me know where. The one example they have has the same issue we have :-(.
This worked:
...
local DISPLAY_WIDTH = sys.get_config_int("display.width")
local DISPLAY_HEIGHT = sys.get_config_int("display.height")
...
w = w / (w / DISPLAY_WIDTH)
h = h / (h / DISPLAY_HEIGHT)
...
This does not
...
local scale = window.get_display_scale()
w = w / scale
h = h / scale
...
From my view, the bee.script should be like that, i.e.
We need to use screen_x and screen_y instead of x and y.
We don’t need to get the display scale, because screen_x/y is in the same unit as the values from window.get_size().
-- function to convert screen (mouse/touch) coordinates to
-- world coordinates given a camera component
-- this function will use the camera view and projection to
-- translate the screen coordinates into world coordinates
local function screen_to_world(x, y, z, camera_id)
local projection = camera.get_projection(camera_id)
local view = camera.get_view(camera_id)
local w, h = window.get_size()
-- https://defold.com/manuals/camera/#converting-mouse-to-world-coordinates
local inv = vmath.inv(projection * view)
x = (2 * x / w) - 1
y = (2 * y / h) - 1
z = (2 * z) - 1
local x1 = x * inv.m00 + y * inv.m01 + z * inv.m02 + inv.m03
local y1 = x * inv.m10 + y * inv.m11 + z * inv.m12 + inv.m13
local z1 = x * inv.m20 + y * inv.m21 + z * inv.m22 + inv.m23
return x1, y1, z1
end
function init(self)
-- send input events to this script
msg.post(".", "acquire_input_focus")
end
function on_input(self, action_id, action)
if action_id == hash("touch") and action.pressed then
-- convert mouse/touch screen position to world position
local worldx, worldy = screen_to_world(action.screen_x, action.screen_y, 0, "#camera")
local world = vmath.vector3(worldx, worldy, 0)
go.animate(".", "position", go.PLAYBACK_ONCE_FORWARD, world, go.EASING_LINEAR, 0.5, 0)
end
end
The only place where we need display scale is for zooming in the camera. For example, here’s how to get a fixed fit projection with display scale taken into account using the camera component:
local DISPLAY_WIDTH = sys.get_config_int("display.width")
local DISPLAY_HEIGHT = sys.get_config_int("display.height")
--- Adjust camera zoom so that the original area covered by
-- display width and height of game.project is always visible
local function fixed_fit_projection(global_zoom, camera_url)
local w, h = window.get_size()
local scale = window.get_display_scale()
local zoom = global_zoom * math.min(w / scale / DISPLAY_WIDTH, h / scale / DISPLAY_HEIGHT)
go.set(camera_url, "orthographic_zoom", zoom)
end
--- Usage example:
function update(self, dt)
fixed_fit_projection(0.6, "#camera")
end
The web page needs to be updated as well (maybe that will happen later and is already planned, but wanted to make sure this doesn’t get missed). Thank you.
and less code that looks much nicer.
